# The Multivac — Evaluation Report

**Evaluation ID:** EVAL-20260315-031910
**Date:** Mar 15, 2026
**Category:** code
**Question ID:** EVAL-20260315-031910

---

## Question

Write a Python function that returns the second largest value from a list of integers. Handle edge cases.

---

## Winner

**Qwen 3 32B** (openrouter)
- Winner Score: 9.66
- Matrix Average: 9.26
- Total Judgments: 79

---

## Rankings

| Rank | Model | Provider | Avg Score | Judgments |
|------|-------|----------|-----------|----------|
| 1 | Qwen 3 32B | openrouter | 9.66 | 8 |
| 2 | Kimi K2.5 | openrouter | 9.52 | 9 |
| 3 | Granite 4.0 Micro | openrouter | 9.38 | 8 |
| 4 | Gemma 3 27B | openrouter | 9.34 | 8 |
| 5 | Qwen 3 8B | openrouter | 9.24 | 8 |
| 6 | Llama 4 Scout | openrouter | 9.19 | 8 |
| 7 | Llama 3.1 8B | openrouter | 9.13 | 8 |
| 8 | Devstral Small | openrouter | 9.11 | 7 |
| 9 | Mistral Nemo 12B | openrouter | 9.09 | 7 |
| 10 | Phi-4 14B | openrouter | 8.96 | 8 |

---

## 10×10 Judgment Matrix

Rows = Judge, Columns = Respondent. Self-judgments excluded (—).

| Judge ↓ / Resp → | Qwen 3 32B | Devstral Small | Gemma 3 27B | Llama 4 Scout | Phi-4 14B | Granite 4.0 | Qwen 3 8B | Mistral Nemo | Llama 3.1 8B | Kimi K2.5 |
|---|---|---|---|---|---|---|---|---|---|---|
| Qwen 3 32B | — | · | 9.3 | 8.6 | 9.0 | 9.8 | 9.0 | · | 9.3 | 9.6 |
| Devstral Small | 9.4 | — | 9.4 | 9.6 | 8.8 | 9.4 | 9.6 | 9.6 | 9.4 | 9.8 |
| Gemma 3 27B | 9.8 | 9.6 | — | 9.4 | 9.4 | 9.2 | 9.1 | 9.2 | 9.6 | 9.8 |
| Llama 4 Scout | 10.0 | 9.3 | 9.6 | — | 8.4 | 9.3 | 9.6 | 9.3 | 9.3 | 9.8 |
| Phi-4 14B | 9.8 | 8.4 | 9.6 | 9.6 | — | 9.4 | 9.6 | 9.6 | 9.0 | 9.8 |
| Granite 4.0 | 9.3 | 8.6 | 8.8 | 8.6 | 8.6 | — | 8.6 | 8.6 | 8.4 | 8.8 |
| Qwen 3 8B | 9.8 | 9.8 | 9.6 | 9.4 | 9.6 | 9.6 | — | 9.2 | 9.3 | 9.8 |
| Mistral Nemo | 10.0 | 8.7 | 9.3 | 9.3 | 9.4 | 8.9 | 9.3 | — | 8.7 | 9.1 |
| Llama 3.1 8B | 9.3 | 9.4 | 9.3 | 9.1 | 8.4 | 9.4 | 9.2 | 8.2 | — | 9.3 |
| Kimi K2.5 | · | · | · | · | · | · | · | · | · | — |

---

## Methodology

- **10×10 Blind Peer Matrix:** All models answer the same question, then all models judge all responses.
- **5 Criteria:** Correctness, completeness, clarity, depth, usefulness (each scored 1–10).
- **Self-judgments excluded:** Models do not judge their own responses.
- **Weighted Score:** Composite of all 5 criteria.

---

## Citation

The Multivac (2026). Blind Peer Evaluation: EVAL-20260315-031910. app.themultivac.com

## License

Open data. Free to use, share, and build upon. Please cite The Multivac when using this data.

Download raw JSON: https://app.themultivac.com/api/evaluations/EVAL-20260315-031910/results
Full dataset: https://app.themultivac.com/dashboard/export